Searched refs:post_training_quantize (Results 1 – 14 of 14) sorted by relevance
164 Flag("post_training_quantize", parsed_flags.post_training_quantize.bind(), in ParseTocoFlagsFromCommandLineFlags()165 parsed_flags.post_training_quantize.default_value(), in ParseTocoFlagsFromCommandLineFlags()269 READ_TOCO_FLAG(post_training_quantize, FlagRequirement::kNone); in ReadTocoFlagsFromCommandLineFlags()
174 Arg<bool> post_training_quantize = Arg<bool>(false); member
176 // DEPRECATED: Please use post_training_quantize instead.191 optional bool post_training_quantize = 26 [default = false]; field
452 params.quantize_weights = toco_flags.post_training_quantize(); in Export()
179 if flags.post_training_quantize:180 converter.post_training_quantize = flags.post_training_quantize
245 post_training_quantize=False, argument333 toco.post_training_quantize = post_training_quantize
498 self.assertFalse(quantized_converter.post_training_quantize)500 quantized_converter.post_training_quantize = True501 self.assertTrue(quantized_converter.post_training_quantize)
9 `--post_training_quantize` flag to your original tflite_convert invocation. For16 --post_training_quantize
87 converter.post_training_quantize = kwargs["post_training_quantize"]292 converter, post_training_quantize=True, **kwargs)
161 model_coverage.test_keras_model(keras_file, post_training_quantize=True)
63 converter.post_training_quantize = True # If post training quantize is desired.97 converter.post_training_quantize = use_post_training_quantize299 converter.post_training_quantize = use_post_training_quantize
245 simply enable the ‘post_training_quantize’ flag in the TensorFlow Lite251 converter.post_training_quantize=True
143 * `--post_training_quantize`. Type: boolean. Default: False. Boolean
178 * `post_training_quantize` - Deprecated in the 1.X API